Your First Agent — Full Walkthrough
This is the extended version of the quickstart, covering every configuration option.
Creating an agent
Section titled “Creating an agent”- If this is your first agent, start from the Console home — it offers Describe it, Start from a template and Start blank
- From your second agent on, go to Console → Agents and click New agent
- Fill in the Basic Info:
- Name — something descriptive, e.g., “Support Bot” or “Dev Assistant”
- Description — a short note about what this agent does (internal reference only)
- Type — see Agent Types → to pick the right one
Configuring the agent
Section titled “Configuring the agent”Instructions tab
Section titled “Instructions tab”The instructions are the most important part. Define:
- Who the agent is — its role and identity
- What it should do — the tasks it handles
- How it should behave — tone, format, what to do when it’s unsure
See Writing Agent Instructions → for detailed guidance and examples.
Capabilities tab
Section titled “Capabilities tab”Capabilities are the built-in abilities of the agent runtime — turn on only what this agent needs:
- Run code / shell in a sandbox — file operations, git, and terminal commands in an isolated container
- Drive a real browser (Chromium) — navigate sites, extract data, and drive web apps
- Web search — look up up-to-date information from the web
- Computer use — vision-based clicks, typing, and scrolling for web UIs (delegated to a sub-agent)
- Voice — real-time voice conversations
- Media generation — create images, video, audio, and diagrams
Internal agents include sandbox and browser capabilities by default; customer-facing agents do not receive sandbox or browser tools. See Sandboxes → and Computer Use & Browser →.
Integrations tab
Section titled “Integrations tab”Select which integrations this agent can use. Only enable what’s needed — this helps keep the agent focused and prevents accidental writes.
Knowledge tab
Section titled “Knowledge tab”Attach knowledge collections the agent should be able to search. You can attach multiple knowledge collections to one agent.
Skills tab
Section titled “Skills tab”Enable built-in skills for common task types: Coding, Research, Customer Support, DevOps. See Skill Library →.
Settings tab
Section titled “Settings tab”| Setting | Description |
|---|---|
| Execution Mode | Fast (fast, lower cost) or Deep (thorough). Defaults to automatic. |
| Monthly Credit Limit | Cap the maximum credits this agent can spend per month. |
| Memory | Enable persistent memory — the agent remembers context across separate conversations. |
| Language | Response language override. Defaults to the user’s input language. |
Testing the agent
Section titled “Testing the agent”Click Start chat on the agent’s page to talk to it in the chat workspace before you deploy it. Your very first agent offers the same thing as Open in workspace on the Console home.
Things to test:
- Does it follow the instructions?
- Does it use the knowledge correctly?
- Does it use tools when needed?
- What happens when you give it an edge case or an off-topic request?
Deploying the agent
Section titled “Deploying the agent”Once you’re happy with the agent in testing, deploy it:
- Web Widget — embed a chat bubble on your website
- Telegram — connect a bot via BotFather
- Slack — add as a workspace bot
- WhatsApp — via WhatsApp Business API
- REST API — call from your own application
- Flows — trigger the agent automatically from events
Monitoring runs
Section titled “Monitoring runs”After you deploy, go to the Runs tab on the agent page to see:
- A history of all tasks the agent has run
- What tools it called and what they returned
- The output it produced
- Credit usage per run
This is your primary tool for debugging and improving the agent over time.